The role of executive function, specifically inhibitory control and generativity, in symbolic play was investigated in 20 children aged 48-89 months. Assessment of inhibitory control was via the Sun-Moon Stroop task, and generativity was assessed with the Semantic Fluency task, as well as a new object substitution task which required children to generate as many uses of toys as possible. Symbolic play ability was assessed under both structured conditions, using the Test of Pretend Play (Lewis & Boucher, 1997), and during free play. The results indicated that the ability to inhibit prepotent responses was associated with children's symbolic play skills, even after controlling for mental age. In contrast, generativity scores on both tasks were not correlated with symbolic play, indicating that only some aspects of executive function are implicated in symbolic play. The validity of the tasks used to measure executive function is discussed. (Contains 3 tables and 1 footnote.)